Nut backed off ball caused accident

[iamalienru]

I was heading up to santa cruz ca. in my 40' american eagle I had my motorcycle in tow on a utility trailer. Before I left I had tighten the nut on the ball with an air impact driver the kind that you take lug nuts off a car or truck it was tight lock washer in place. safety chain hooked up all was good. traveling down the 101 near north Hollywood ca. I saw the trailer jumping in my rear camera as I slowed the trailer let loose breaking the safety chain and hitting another car. what a mess. I have never had this happen before but the nut from the ball was laying there on the freeway. now I am having insurance issues on the trailer and coach it was the vacation from hell and it's not over. The only type of ball I will use from now on is one with a cotter pin going through the bottom of the shaft. Uhaul carries them. Sorry for the sad story but i hope this will help others. I still do not understand why that nut backed off

[calewjohnson]

BurbMan wrote:

BenK wrote:
Fastening science stuff...cotter pin & castle nut are for different applications

They are for 'not' too tight, but just enough for bearings or some such that needs
to be allowed to 'move'.

Clamping force is NOT the goal, but maintained 'gap' is

Yes that is their role in that specific application but not always.

Look at the head bolts on a radial aircraft engine. All of the bolt heads have a hole drilled parallel to the top and from flat to flat, like this:


Two step fastening process: First the bolts are torqued to spec, second a long loop of safety wire is run through the head of the bolt, the two ends twisted as a pair and then onto the next bolt like this:



On a radial aircraft engine this goes all the way around the perimeter of the engine and includes all of the bolts fastening the head on.

In the case of studs, the nuts are castled and wired in the same fashion.

This NOT a replacement for the correct torque but in addition to. insures that the bolts don't come loose from extreme vibration, cold temps at altitude, etc.

So back to the hitch ball....rather than relying solely on torque to hold the fastener in place, why not use a safety mechanism? Instead of a torque-only spec of 400+ ft/lbs, why not a more attainable 150 ft/lbs with a castle nit and cotter pin?

Also referred to as Safety Wiring. If done properly, when a bolt loosens and tries to back off, it will be tightening the bolt next to it...not just for radial engines, but used throughout aviation period. I fly now, use to be a maintainer, I will tell you that CDI/QA personnel pay especially close attention to the way these procedures are done before you get thrown off the pointy end of a carrier.

Cale

[69_Avion]

I'm glad nobody was hurt. Don't assume that it came loose on it's own. Someone could have messed with it.
Having built equipment trailers on and off since the mid 1970s, I have never had one come loose on it's own. I also went with pintle hitches on those larger trailers back in the late 1970s. A small utility trailer would probably have a 2" ball. Using a cotter pin would certainly make you feel better after what you have been through.

[WyoTraveler]

Dave H M wrote:

WyoTraveler wrote:
Tip the hitch sideways and slide it into the receiver. You don't even need a torque wrench. calculate your weight and length of pipe breaker bar you need and stand on it. 3/4" nut 160 pounds, 1" nut 250 pounds, 1 1/4" nut 450 pounds ft pounds. I personally wouldn't trust any air wrench for torque.

Wow we have some home grown ingenuity how to here. That is what I did when I "torqued" the tractor axle nut on. I was not smart enough on the hitch ball and sat on the ground and had both feet up against the bumper and then proceeded to see if i could turn the breaker bar into a long bow. :S

Yup, learn all that stuff working around farm equip. Just retired final time raising hay and cows in WY.:B

[Passin_Thru]

I picked up a single-man, 40 ft reach, manlift for a friend and hauled it to a site as a volunteer for a town mural. When I went back to pick it up, I chained it, crossing the chains properly. On an onramp, it broke the hitch and as I slowed down the whole basket came through my tailgate. The whole project cost me $1450.00 because my deductable was $1000 and the tailgate $1100. A properly torqued ball could break or a hitch could break and cause the same problems. Quit judging the poor guy, he shared a lesson. Learn from it!
I went to torquing school for Aviation Technicians and they teach you it's to stretch the bolt properly to it's maximum holding strength. They also use SS safety wire to hold a large aircraft engine to the pylon and I did many DC 10s, MD 11s and C17s as a Flight Test Tech for McDonnell Douglas.

[Turtle_n_Peeps]

So back to the hitch ball....rather than relying solely on torque to hold the fastener in place, why not use a safety mechanism? Instead of a torque-only spec of 400+ ft/lbs, why not a more attainable 150 ft/lbs with a castle nit and cotter pin?

For the same reason we don't put a castle nut or cotter pin in a rod bolt or head bolt. A properly torqued nut is more than just holding a nut on.

If it was just for holding the nut on I could skip all the rod torqueing nonsense and just use Loc-Tite on them.

If you don't torque a rod bolt properly the rod will fail in short order. And not because the nut fell off.

If you do not torque a hitch ball to the proper spec it "can" fail for the same reason as the rod.

The reason we torque fasteners has VERY little to do with them coming loose.

[BurbMan]

BenK wrote:
The more I read on these forums...the more concerned get about the setup next to me going 65MPH or more...

LOL it's not that bad!

Thanks for explaining, that makes sense. Only asking because I'm not an engineer and don't have the same level of understanding on what goes into these design decisions. So need the clamping force to prevent movement of the ball shank in the hole, not just to keep the nut on. Makes sense.

So my revised recommendation on the castle nut would be to torque to 450 ft/lbs THEN cotter it!

You should move to KY where trailers for non-commercial use don't need to be registered (TTs and 5ers require tags) and there is no MV inspection of any kind, forget about a smog check....you see some interesting stuff on the road down here....

[BenK]

Not everyone will go to a pintle, nor weld, nor anything else...though there
will be some

Hope brought this down to a level for some here to understand what
the heck this is for an designer

Here is the formula to figure out the clamping force. Clamping force is
calculated according to the dynamic loads and shock loads the
designer worked out for the worst case...plus margin. Once they
have the clamping force needed, then the process of selecting a
bolt (shank) size, or bolts. Along with grade and another whole
discussion on the selection of grade...cascading failure is one
where you want the lower grade on the part you wish to fail first, etc

T / KD = P

P=clamping force
D=shank dia or nominal thread dia
K=Constant and is the co-efficient of friction on the bolt/nut
T=Torque

On the constant and is the friction between the mating surfaces that the torque
must over come *BEFORE* the fastener sees any clamping force... 0.16 is a very
smooth and greased mating surfaces...rusty and rough surfaces would be around 0.75, or more (these numbers are IIRC, as it has been decades since the last
time I've calculated a fastener for a design)

Chose 1" shank dia for ease of math



Two thoughts on this.

The ball instructions of 450 ft/lbs
Theory of 150 ft/lbs & cotter/castle will do

Just the numbers for now

450 ft/lb / 0.5 x 1 = 900 lbs of clamping force

150 ft/lb / 0.5 x 1 = 300 lbs of clamping force, along with a cotter or castle


But there is more than just the clamping force, though the main component.


There is the hole that the ball shank goes through. It has tolerance, meaning
bigger than the shank by some designed in margin to allow the shank to easily
pass through...but that is part of the problem...that tolerance

If the hole is sized as a press fit, it will still have some movement of the
shank in that 'tight' hole

Movement is a VERY bad thing for this application

So if the clamping force is 1/3 of the designers spec, then the shank will
most likely move around in that hole...even it it has a cotter and castle too

Banging and over time will work the metal of both the shank and hole.

Moving like that to bang will also allow the nut to back off, but okay...the
cotter will stop that via interference...but the castle *WILL* back off, unless
it too have a cotter in the castles

Here is the biggie over and above the previous...the shank will be allowed to
lean and/or bend because it does NOT have the preload spec'd in by the designer

Since not enough preload for that application (meaning the static and dynamic
loads the designer worked to)...the loading will pull/push/etc the ball that
will then lever the shank.

All of that movement *WILL* cause other things to wear faster. Like the coupling
and it's latch


If tightened to the spec of 450 ft/lbs, which creates 900 pounds of clamping
force...it should not move when exposed to expected forces

Must qualify that...always say things designed are designed not for the good
days out there, but for the worst day out there when Mr Murphy crosses your path

'Maybe' the cotter/castle will keep the nut on, but with all of the increased
wear and tear...will the coupler hang onto the ball? Will the ball bend over
enough to create a condition where the couple dome recess will allow the ball
to move out of?

The more I read on these forums...the more concerned get about the setup next
to me going 65MPH or more...

[hammer21661]

You know we are all kind of dumb everyone that has commented in this thread. I have the best solution. Get the triple balls that are welded to the solid square stock. If an adjustable unit is desired the B and W tow and stow is awesome. These balls are what I use from the receiver. Better yet is a pintle. Although we don't want accidents if you use a commercially built one that fails it is better then a nut coming loose,from lack of attention, at least you can play the blame game.

[BurbMan]

BenK wrote:
Fastening science stuff...cotter pin & castle nut are for different applications

They are for 'not' too tight, but just enough for bearings or some such that needs
to be allowed to 'move'.

Clamping force is NOT the goal, but maintained 'gap' is

Yes that is their role in that specific application but not always.

Look at the head bolts on a radial aircraft engine. All of the bolt heads have a hole drilled parallel to the top and from flat to flat, like this:


Two step fastening process: First the bolts are torqued to spec, second a long loop of safety wire is run through the head of the bolt, the two ends twisted as a pair and then onto the next bolt like this:



On a radial aircraft engine this goes all the way around the perimeter of the engine and includes all of the bolts fastening the head on.

In the case of studs, the nuts are castled and wired in the same fashion.

This NOT a replacement for the correct torque but in addition to. insures that the bolts don't come loose from extreme vibration, cold temps at altitude, etc.

So back to the hitch ball....rather than relying solely on torque to hold the fastener in place, why not use a safety mechanism? Instead of a torque-only spec of 400+ ft/lbs, why not a more attainable 150 ft/lbs with a castle nit and cotter pin?

[Turtle_n_Peeps]

There is really no reason for lock washers or Loctite or welding or cotter pin or smashing threads or any of that at all.

It real simple. Tighten it up to torque spec and you will have no problems with any nut backing off. If you don't have to tools to do this(most don't)then please have a shop do this for you.

If you don't get the fastener to torque spec you can have other problems that you don't want.

[BenK]

Fastening science stuff...cotter pin & castle nut are for different applications

They are for 'not' too tight, but just enough for bearings or some such that needs
to be allowed to 'move'.

Clamping force is NOT the goal, but maintained 'gap' is

Curious why folks do NOT trust the designers who wrote the specification for the
torque on 'their' design

Know and understand that am in the vast minority because this is technology in
discussion...but...why don't you folks believe the instructions?

Would love to have some of you folks hands on build any engine without using
a torque wrench...components and machining costs out of your own pockets (your
skin in the game)

Doesn't matter small block, big block, diesel or even a 2 cycle chain saw ICE

Then ask it to go +100K miles under very hard conditions


Oh well...moving on...and get off this soap box...

[BurbMan]

Surprised all of these balls don't come with a castle nut and a cotter pin set up like on a wheel bearing. Instead of using a cotter as last resort down at the end of the threads, use a castle nut and drill the hole for the cotter up by the nut. You wouldn't need to tighten the bejesus out of them at 400+ ft/lbs if the nut was captive and couldn't back off like it is on a wheel bearing.

[beergardens]

I am thinking what may have happened is that the OP may have over tightened the nut, causing the lock washer under it to spread and break. After a little while on the road, leverage on the ball would have caused a piece of lock washer to spit out, leaving the nut very loose on the ball's shank. The main point I am making is that even if you use a torque wrench, have a good look at the lock washer after. I've sure seen a lot of them spread and try to come out from under the nut.

[VintageRacer]

With respect to the chains, on almost all my trailers I replace the chain with Gr 70 DOT rated chain and use DOT rated hooks with spring closures. Piece of mind. Light trailers in particular are often sold with inadequate chains and bent wire hooks that I don't like at all.

Brian

[Raften]

Get a big wrench that fits the nut from a large fishing supply outfit, Cabela's Bass Pro Shop, ect. They are a good price and cheap insurance.